Kidney International, Vol. 42 (1992), pp. 947—950

Peritoneal defense in continuous ambulatory versus continuous cyclic peritoneal dialysis CAROLA W.H. DE FIJTER, HENRI A. VERBRUGH, LIEM P. OE, EDITH D.J. PETERS, JAN VAN DER MEULEN, AB J.M. DONKER, and JAN VERHOEF Department of Internal Medicine, Free University Hospital, Amsterdam, and Department of Microbiology, State University of Utrecht, Utrecht, The Netherlands

Peritoneal defense in continuous ambu'atory peritoneal dialysis versus continuous cyclic peritoneal dialysis. Several centers have reported a lower rate of peritonitis among adult patients on Continuous cyclic peritoneal dialysis (CCPD) as compared to those undergoing continuous

randomized study comparing CAPD and Y-connector (CAPD-Y) with CCPD also suggest a lower peritonitis incidence

among CCPD-treated patients [4]. The lower peritonitis inciambulatory peritoneal dialysis (CAPD). Preliminary results of our dence observed in CCPD as compared to CAPD has primarily ongoing prospective randomized study comparing CAPD-Y with CCPD been associated with the fewer (dis)connections needed in also suggest a lower peritonitis incidence among CCPD-treated paCCPD. However, local defense mechanisms may also play an tients. To investigate whether the two dialysis regimens could result in important role in the prevention of and recovery from bacterial differences in local host defense, we studied peritoneal macrophage (PMO) function and effluent opsonic activity in eight patients estab- peritonitis; in this respect ingestion and killing of microorganisms by peritoneal macrophages are thought to be essential [5]. lished on CAPD-Y matched with eight chronic CCPD patients. Since short and long dwell times are inherent to both dialysis modalities, and Because the dialysis regimens inherent to CAPD and CCPD, we previously found that dwell time has an impact on PMO function and respectively, differ primarily in their dwell time periods precedeffluent opsonic activity, patients were studied after both a short (4 hr) ing exchange procedures, we studied PMO function and effluent and a long (15 hr) dwell time. In both patient groups PMO phagocytic capacity increased significantly with dwell time (39 3.3% at 4 hr vs. opsonic activity after both a short (4 hr) and a long (15 hr) dwell 58 4.2% at 15 hr in CAPD patients, and 40 3.9 vs. 72 3.3% in time in eight CAPD-Y patients matched with eight CCPD CCPD patients; P < 0.01), as did PMO peak chemiluminescence patients. response (31 4.9 vs. 77 7.2 counts min'/l04 cells in CAPD, and 22 3.9 vs. 109 21.2 counts min/l04 cells in CCPD; P < 0.01) and effluent opsonic activity (41

7.6 vs. 73

Methods

5.8% in CAPD and 39

6.2 vs. 70 5.9% in CCPD; P < 0.01). However, no significant difference was found in either variable between CAPD and CCPD patients when dwell times were equal. In conclusion, no differences

Patients and study design

Eight stable CAPD-Y patients [4 men and 4 women, median age 55 years (35 to 76) and on CAPD for a mean SD period of 26.6 14 months] matched with eight CCPD patients [5 men, 3 women, median age 53.5 years (37 to 73) and on CCPD for a defense. The improvement in peritoneal defenses may, in part, be mean SD of 25.8 12.9 months] were studied. The etiologies responsible for the lower peritonitis incidence observed among CCPDof end-stage renal failure were glomerulonephritis (2 patients), treated patients. interstitial nephritis (2 patients), nephrosclerosis (2 patients), diabetic nephropathy (2 patients) for the CAPD-Y group, and were observed in PMO function or effluent opsonic activity between matched CAPD-Y and CCPD patients when dwell times were equal. In both patient groups prolongation of dwell time enhanced PMO function as well as effluent opsonic activity, thereby providing a better host

glomerulonephritis (2 patients), interstitial nephritis (3 pa-

Peritoneal dialysis has become an increasingly applied treat- tients), nephrosclerosis (2 patients) and diabetic nephropathy (1 ment modality for patients with end-stage renal disease. How- patient) for those on CCPD. At time of study entry, there was

ever, bacterial peritonitis still is the major complication of no evidence of peritonitis (that is, no symptoms and clear continuous ambulatory peritoneal dialysis (CAPD), resulting in relatively high morbidity and drop-out rates [1]. Several centers

applying continuous cyclic peritoneal dialysis (CCPD) on a

effluents with less than 100 WBC/mm3 for at least 4 weeks prior

to participation). In a randomized cross-over setting each patient performed the first exchange (2 liters of Dianeal®,

large scale have reported significantly lower rates of peritonitis containing 2.27% glucose, Baxter Ltd., Thetford, UK) of study among adult CCPD patients as compared to those undergoing day one after a dwell time of 4 or 15 hours. The next day, the CAPD [2, 3]. Preliminary results of our ongoing prospective patients who had performed the first exchange of study day one after a dwell time of 4 hours, exchanged their first bag (2 liters of Dianeal®, containing 2.27% glucose) after a dwell time of 15 Received for publication August 22, 1991 and in revised form April 15, 1992 Accepted for publication April 19, 1992

1992 by the International Society of Nephrology

hours and vice versa. The peritoneal effluents were collected and studied for total WBC and differential count, IgG and C3 concentration, as well as opsonic activity for Staphylococcus aureus. Functional studies of PMO isolated from the effluents

947

948

de Fzjter

et a!: Peritoneal defense in CAPD vs. CCPD

included both their phagocytic capacity and their ability to lation counting [9]. Phagocytosis was expressed as percentage uptake of total radioactivity added, which was determined in a approved by the Ethical Review Committee of the Free Uni- separate vial [7]. versity Hospital and informed consent was obtained from all Assessment of effluent opsonic activity for S. aureus patients. The uptake of radiolabeled S. aureus, preopsonized in undiPhagocytic cells luted effluent, by PMN, was taken as a measure of effluent Peritoneal macrophages (PMO) were isolated from the efflu- opsonic activity (see phagocytosis assay). Bacteria, preopents as previously described [61. Final resuspensions were sonized with 5% HPS, were used as controls in each experimade in Hank's balanced salt solution containing 0.1% gelatin ment. The concentrations of IgG and C3 in effluent were (GHBSS) to a concentration of 5 x 106 PMO/ml. Total and measured nephelometrically using a Beckmann protein assay differential cell counts were performed on all samples. Viability (Beckmann, Mijdrecht, The Netherlands). was assessed by the trypan blue exclusion test. Test-control Chemiluminescence assay phagocytes were peripheral blood polymorphonuclear leucocytes (PMN) isolated from 50 ml sterile, heparinized venous The microchemiluminescence assay described by Mills, blood of healthy donors using dextran sedimentation and a Rholl and Quie [10] was modified and used to measure the Ficoll gradient [7], and were also used at a concentration of 5 X oxidative metabolic responses of PMO obtained after different mount a respiratory burst upon stimulation. The study had been

dwell times. Amplification of the chemiluminescence signal was obtained by adding luminol (0.03 mmol/ml in GHBSS; Sigma Bacteria Chemical, St. Louis, Missouri, USA). Chemiluminescence mixA clinical isolate of S. aureus, obtained from a CAPD patient tures, in transparent plastic scintillation vials, contained 800 sl with peritonitis was used. Bacteria were radiolabeled by grow- GHBSS, 100 d PMO or PMN (1 x l0 cells/mi), and 70 pi ing them in 10 ml of Mueller-Hinton broth (Difco Laboratories, luminol. Phagocyte chemiluminescence was activated by addUSA) containing 20 tCi 3H-adenine (ICN, USA) in a shaking ing 100 d phorbol myristate acetate (PMA; 50 ng/ml in GHBSS; incubator for 18 hours at 37°C [10]. Subsequently, bacteria were Sigma Chemicals). Backgrounds were measured in separate washed three times with phosphate-buffered saline (PBS) and vials devoid of stimuli (100 d GHBSS instead of PMA). After resuspended in GHBSS to a concentration of 5 X 108 cfu/ml [8]. stimulation, counts were obtained every two minutes for 30 minutes at 37°C. Results were recorded as the peak number of Opsonins and opsonization procedure counts per minute per lO cells after subtracting background A pool of human serum (HPS) obtained from ten healthy counts. (HBs-antigen and HIV antibodies negative) donors, stored at Data analysis —70°C, was used as source of opsonins. Immediately prior to use, serum was thawed and diluted to a final concentration of Paired two-tailed Wilcoxon signed rank tests were used for 5% in GHBSS. Suspensions of bacteria (0.1 ml containing the comparison of the matched CAPD-Y versus CCPD patients approximately 5 x io microorganisms) were mixed with 0.9 ml at a given dwell time and for the comparison of short versus 106

cells/ml.

of 5% HPS and incubated in a shaking water bath for 30 minutes

long dwell time within either dialysis modality, where the

at 37°C. Opsonization was stopped by adding 2.5 ml ice-cold patients served as their own controls. A probability of 5% was PBS. Subsequently, the suspensions were centrifuged at 3000 chosen as level of significance. rpm for 15 minutes, the supernatants discarded, and the bacterial pellets resuspended in 1 ml GHBSS and kept at 4°C until use (preopsonized bacteria). Cell-free effluents were also stored at —70°C until they were

likewise used as source of opsonins to test their opsonic

Results

The effluent total white cell count from both CAPD and CCPD patients increased significantly with dwell time; at 4 hours and 15 hours the WBC was 2.6 0.46 versus 6.7 0.64

activity. The effluents, however, were used undiluted. In addi- x 10 cells/liter, respectively, in CAPD patients, and 2.2 0.38 tion part of the serum and effluents were heated (56°C for 30 versus 6.2 0.66 x 10 cells/liter, respectively, in CCPD mm) to deplete them of heat-labile opsonins prior to their use. patients (P < 0.01). However, there was no change in cellular composition of the effluents; the overall white cell differentiaPhagocytosis assay tion was 89% PMO (range 82 to 93%), 7% lymphocytes (4 to The uptake of preopsonized S. aureus by PMO was deter- 9%), 3% PMN (2 to 5%) and 1% mesothelial cells (0 to 2%). mined using an assay that has been described in detail [7]. In Viability was always over 90%. PMO obtained after a 15-hour brief, 100 d of bacteria preopsonized with 5% HPS was mixed dwell time period revealed a significantly better uptake of S. with 100 pi of PMO in a final bacteria to phagocyte ratio of 10:1. aureus as compared to PMO obtained after a dwell of 4 hours, In parallel, donor PMN incubated with the same preopsonized whereas no significant difference was found between PMC cocci were run as control. Phagocytosis was allowed to proceed phagocytic capacity of CAPD and CCPD patients when dwell for 60 minutes in a shaking incubator at 37°C. Phagocytosis was times were equal (Table 1). stopped by adding 2.5 ml of ice cold PBS. Non-phagocyteThe same held true for PMO chemiluminescence response, associated bacteria were removed by centrifugation (1200 rpm reflecting the ability of PMO to mount a respiratory burst, that for 5 mm), washing three times with ice cold PBS followed by is, to produce toxic oxygen radicals upon stimulation with three cycles of centrifugation (1200 rpm for 5 mm). The phago- PMA: PMO derived from both CAPD and CCPD patients after cyte-associated radioactivity was determined by liquid scintil- a dwell time of 15 hours revealed a significantly higher peak

de Fjjler et at: Peritoneat defense in CAPD vs. CCPD Table 1. The effect of dwell time on PMO function in CAPD and CCPD Dwell time (hr)

PMO function

% Uptake S. aureus Peak CL response

CAPD

4

39

3.3

15

58

4.2a

4

31

4.9

15

77

7.2a

CCPD 40

3.9

22 109

3.9

72 33a 2l.2'

Data are mean SEM. Abbreviation is: CL, chemiluminescence in counts per mm per io cells. a p < 0.01 vs. 4 hr; Control phagocytes (PMN; N = 8) phagocytized 86 1.97% of S. aureus, and their chemiluminescence response was

954 64.3 counts min'/104 cells

Table 2. Effluent opsonic activity and IgG- and C3 levels after different dwell times Dwell time (hr)

Total opsonic activity (OA) % Heat-stable OA %

4 IS

4 15

Effluent IgG gluier

4 15

Effluent C3 glliter

4 15

CAPD 41 73 33

39

6.2

5.8k

70 5.9

5.9

33 41

4.4

7.6

48 51b 0.010 0.070a